AI

CH-4 – Mean Frame Extractions – Part-3

Posted on Author Ali Imran Comment(0)

Important Note: This article is part of the series in which TechReport.us discusses the theory of Video Stream Matching.

4.3.1 Algorithm 1

The Mean frame extraction process serves a dual purpose, operating in both Source and Target Videos. The mechanism remains consistent for both cases, employing the same algorithm.

This algorithm (Algorithm 1) is designed to work with videos in AVI format, focusing on the fundamental task of extracting mean frames.

Input:

  • AVI Video

Output:

  • BMP Images stored on the hard disk

Working:

  1. The algorithm takes a video and stores its first and last frames as they are on the hard drive.
  2. A defined percentage of frames (10% in this system) is extracted from the video, referred to as temporary frames.
  3. The algorithm then processes each temporary frame, calculating its histogram.
  4. The comparison of histograms is conducted using the KS Test. If two frames have the same distribution, the second one is neglected; otherwise, the second temporary frame is saved as a mean frame.
  5. This process is repeated until all frames are treated.

Algorithm Steps:

  • Step 0: Start
  • Step 1: Input a video.
  • Step 2: Store the video on the hard disk.
  • Step 3: Set the total number of frames (NumOfFrame) and define the percentage of frames to skip (TempSkip = 10%).
  • Step 4: Loop for i = 1:TempSkip:NumOfFrame
  • Step 5: Extract temporary frames from the video.
  • Step 6: Store temporary frames on the hard disk.
  • Step 7: If the loop does not end, go to Step 4.
  • Step 9: Store the first frame of the video on the hard drive.
  • Step 10: Initialize CountMeanFrame = 0.
  • Step 11: Increment CountMeanFrame by 1.
  • Steps 12-20: Perform histogram calculations and KS Test for temporary frames.
  • Step 21: If the loop does not end, go to Step 13.
  • Step 22: Store the last frame on the hard disk and increment the counter.
  • Step 23: Store in the file the number of frames selected as mean frames and their names.
  • Step 24: Stop
Ali Imran
Over the past 20+ years, I have been working as a software engineer, architect, and programmer, creating, designing, and programming various applications. My main focus has always been to achieve business goals and transform business ideas into digital reality. I have successfully solved numerous business problems and increased productivity for small businesses as well as enterprise corporations through the solutions that I created. My strong technical background and ability to work effectively in team environments make me a valuable asset to any organization.
https://ITsAli.com

Related Articles
AI

CH-6 – References – VSM

Posted on Author Ali Imran

References – VSM
AI

CH-1 – Video Stream Matching – VSM

Posted on Author Ali Imran

The article begins by highlighting the dynamic nature of image processing in the context of advancing information technology. It predicts a future shift from image to video processing due to the increasing prevalence of videos over text-based media. The Video Stream Matcher (VSM) is introduced as a tool for analyzing video data using statistical models like KS Test, Variation, Mean, and Norm.

Key frames extraction is identified as a technique to minimize the vast collection of frames in a video. Four features are extracted from each frame: histogram, edge, slope, and wavelets. The literature review section references research papers and books that inform the implementation of VSM, including studies on similarity analysis of video sequences and key frame extraction.

The key elements of VSM are outlined, including the input of source and test data (videos or images), key frames extraction, feature extraction, and application of statistical models for decision making. The problem statement highlights the need for automated video evaluation systems, particularly in scenarios like security checks on public transportation and video database management in TV stations.

The proposed solution revolves around VSM’s ability to process videos, focusing on key frames extraction, feature extraction, and statistical model application for decision making. The scope of the project encompasses mean frame extraction, feature extraction algorithms, and dual-phase statistical decision making. The organization of the report is structured to cover research findings, implementation details, software functioning, and conclusions with future work discussions.
AI

CH-4 – Edge Feature Extraction – Part-5

Posted on Author Ali Imran

Algorithm 3 extracts edge map features from BMP images. It begins by converting the image to grayscale if necessary and applies the Canny algorithm, returning the image edges. The process involves steps like Gaussian filtering to remove noise, Sobel operator application in horizontal and vertical directions, and angle calculation for each pixel. The resulting edges are thresholded and subjected to double thresholding for refinement. Finally, the processed image with edge maps is returned.

Algorithm 4 operates similarly but focuses on extracting edge angles instead of edge maps. It follows a comparable process, including Gaussian filtering, Sobel operator application, and angle calculation, but stores the occurrences of each angle in tiles of the image, returning an array defining the edge angles.

Leave a Reply